Knowledge catalogue
arXiv:2610.17310v1 Announce Type: new Abstract: Language-model agents are expected to adapt to others whose behavior changes through interaction. Yet adaptation may rely on learned behavioral patterns
arXiv:2610.18760v1 Announce Type: new Abstract: Test-time scaling for large language models (LLMs) can involve generating multiple candidate solutions for selection or revising an earlier attempt. To
arXiv:2610.19882v1 Announce Type: new Abstract: Instruction-driven image editing has reached native 4K, yet how such editors are trained remains undisclosed, and pretraining one from scratch is beyond
arXiv:2610.15153v1 Announce Type: new Abstract: We ask whether a tool-using agent whose sole duty is to keep humans alive can be induced, by delayed prompt injection, to kill. The keep-alive duty is i
arXiv:2610.20012v1 Announce Type: new Abstract: Chemistry evaluation has moved past multiple-choice probes toward open-ended expert problems, yet the benchmarks that define the current frontier still
arXiv:2610.17256v1 Announce Type: new Abstract: Conversational analytics lets people ask questions of records spread across many systems and get answers in plain language; it is useful only if each an
arXiv:2610.19003v1 Announce Type: new Abstract: Abstractive summarization remains challenging across languages, where linguistic quality and factual consistency may diverge, particularly in low-resour
arXiv:2610.14346v2 Announce Type: replace Abstract: Enterprise customer-support agents increasingly rely on Large Language Models (LLMs) to execute bounded, tool-using workflows. We study this problem
arXiv:2610.20417v1 Announce Type: new Abstract: Audits of large language model (LLM) extraction from full-length documents report severe unreliability, leaving two questions open: what the failures ac
arXiv:2610.19996v1 Announce Type: new Abstract: Large language models often produce fluent but factually incorrect output. Gray-box detectors identify such hallucinations from the token-probability st
arXiv:2610.15091v1 Announce Type: cross Abstract: Argument structures represent how individual arguments interact within a discourse. However, leveraging these structures to infer meaningful knowledge
arXiv:2610.19513v1 Announce Type: cross Abstract: Feel-good Thompson sampling (FGTS) augments the posterior of vanilla Thompson sampling with an exploration term that favors reward functions with high
arXiv:2610.14960v1 Announce Type: cross Abstract: Can an autonomous AI agent reverse engineer a binary as well as a skilled human analyst? Do they adopt the same strategy, and can agents scale to real
arXiv:2610.17757v1 Announce Type: new Abstract: Large language model agents can follow complex human instructions to solve real-world tasks. In practice, however, initial instructions are often incomp
arXiv:2610.19796v1 Announce Type: cross Abstract: In Theory-of-Mind (ToM) reasoning, large language models may correctly infer the agent's mental state but fail to use that inference when predicting o
arXiv:2610.19571v1 Announce Type: new Abstract: Real-world engineering drawings are often large-format and information-dense. Answering practical engineering questions therefore requires both fine-gra
arXiv:2610.18092v1 Announce Type: new Abstract: Deep research (DR) agents address open-ended information needs through multi-step reasoning, long-horizon web search, and evidence synthesis. However, e
arXiv:2609.32717v2 Announce Type: replace-cross Abstract: Large Language Model (LLM) alignment is intended to ensure that models remain helpful and safe, but its stability under input distributional s
arXiv:2509.09658v3 Announce Type: replace Abstract: Hallucinations in multimodal large language models (MLLMs), where a model generates content inconsistent with the input image, pose significant risk
arXiv:2610.19920v1 Announce Type: cross Abstract: Molecular design is often formulated as an optimization problem over existing molecular spaces. However, molecular optimization and the discovery of s
arXiv:2610.18755v1 Announce Type: cross Abstract: Reinforcement learning (RL) from unit tests is the common recipe for post-training code models, but a unit test only asks whether the final program pa
arXiv:2610.14912v1 Announce Type: new Abstract: Language models are increasingly used to create synthetic survey samples, also called silicon samples. Decision models are a new class of language model
arXiv:2610.14570v1 Announce Type: new Abstract: Large language models (LLMs) support social simulation, but faithfully reproducing real users requires grounding both structured platform decisions and
arXiv:2610.17584v1 Announce Type: new Abstract: Operational mobile networks are rich in observations but poor in interventions: they continuously generate large volumes of passive telemetry, while con
Relevant to the comprehension challenge I posed in 2014 and my bet with Miles Brundage. We edited Mark Twain's Huckleberry Finn and fed it to 9 AI models to summarize a chapter, twice each. Only Claud
arXiv:2610.15513v1 Announce Type: cross Abstract: Safety-aligned large language models (LLMs) remain vulnerable to adversarial prompts that bypass their safety training. Existing automated red-teaming
arXiv:2610.17665v2 Announce Type: new Abstract: Can vision-language models turn 2D observations into 3D understanding that supports precise action? This requires inferring object geometry, anticipatin
arXiv:2610.19051v1 Announce Type: cross Abstract: Coordinators that act on what they infer about teammates are usually certified by the accuracy of those inferences. We argue that accuracy is the wron
arXiv:2610.14443v1 Announce Type: cross Abstract: Model cards, system cards, and safety reports increasingly document LLM safety behaviors, yet safety evaluations typically start from predefined bench
arXiv:2610.18216v1 Announce Type: cross Abstract: LLM agents can adapt to task feedback by revising the textual context that guides their behavior, without changing their model parameters. However, tu
arXiv:2511.00198v2 Announce Type: replace-cross Abstract: Autoregressive language models are usually trained with next-token prediction under a fixed left-to-right target order. This paper studies tar
arXiv:2610.16155v1 Announce Type: new Abstract: Interactive-agent evaluations commonly rely on standardized LLM-based user simulators that exhibit broadly cooperative behavior, potentially overlooking
arXiv:2610.20104v1 Announce Type: new Abstract: Autoregressive transformers trained on limit order book (LOB) message streams are increasingly proposed as foundation models for market microstructure.
arXiv:2610.18695v1 Announce Type: new Abstract: Kronecker-preconditioned optimizers recompute a cached curvature eigendecomposition only periodically, and implementations and adaptive methods control
arXiv:2610.16324v1 Announce Type: new Abstract: Training-free language-based video anomaly detection (VAD) uses frozen vision--language models (VLMs) and large language models (LLMs) to localize abnor
arXiv:2610.14163v1 Announce Type: cross Abstract: Terminal agents solve tasks in software engineering and scientific research by executing commands in a terminal. Training them via reinforcement learn
arXiv:2610.16745v1 Announce Type: new Abstract: Attention sinks are usually studied after they have formed. We ask whether a measurable hidden-state change appears beforehand and whether changing that
arXiv:2610.14935v1 Announce Type: new Abstract: Pipeline parallelism has become a critical technique for training large language models. To address the throughput limitation of existing sequence-level
arXiv:2409.17397v4 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are increasingly being integrated into various medical fields, including mental health support systems. However,
arXiv:2610.16587v1 Announce Type: new Abstract: Financial candlestick understanding requires locating visual elements, aligning them across trading dates and panels, and combining observations through
arXiv:2606.15320v2 Announce Type: replace Abstract: Multimodal large language models have made rapid progress in video temporal grounding, yet real-world applications routinely require localizing ever
arXiv:2610.17882v1 Announce Type: new Abstract: Flow-based image generators achieve strong semantic synthesis but still produce localized structural distortions, such as malformed hands and limbs. Exi
This is an offline app with 51 hands-on labs for learning how AI actually works, from neurons to agents and more... **A bit of insight** - 51 labs in 6 groups, from the basics (what a neuron is, gradi
arXiv:2610.14112v1 Announce Type: cross Abstract: Calibrated typed-decision ('System One') models return a label and a probability in a single forward pass at a small fraction of the cost of large lan
arXiv:2610.18822v1 Announce Type: new Abstract: Egocentric human videos offer a scalable alternative to robot teleoperation for collecting dexterous manipulation data. However, two challenges limit th
arXiv:2610.17898v1 Announce Type: new Abstract: Sparse autoencoders decompose language model representations into interpretable features, but feature interpretation in this literature remains predomin
arXiv:2610.17724v1 Announce Type: cross Abstract: Long-horizon, semi-verifiable agentic tasks often rely on rubric-based evaluation, but the evidence needed to assess these tasks can be spread across
arXiv:2610.17794v1 Announce Type: new Abstract: Contextual forecasting seeks to improve time series forecasting by incorporating forward-looking information, such as planned events, external condition
arXiv:2610.17370v1 Announce Type: new Abstract: Scientific discovery with language models requires finding the best valid solution within a finite computational budget. We study how feedback from this
I built BitRun because I wanted to run 1-bit ternary LLMs on ordinary laptops without melting my battery, spinning up a dedicated GPU, or wrestling with complex build environments. Right now, BitRun g
arXiv:2610.18572v1 Announce Type: cross Abstract: Automated post-training transforms pre-trained large language models (LLMs) into models capable of handling diverse domains by automatically construct
arXiv:2606.15610v2 Announce Type: replace-cross Abstract: Passing an exact-clone null gate does not entail passing a matched target-selection utility gate. We call a non-tie preference between exact d
arXiv:2508.11624v2 Announce Type: replace Abstract: Low-Rank Adaptation (LoRA) has become a widely adopted technique in text-to-image diffusion models, enabling the personalisation of visual concepts
arXiv:2610.14350v1 Announce Type: cross Abstract: Having demonstrated a strong potential for automating vulnerability discovery, AI agents can also be used to attack software systems at scale. Today's
New impressive work from Microsoft and collaborators. And it's one of my favorite recent topics: RL environments. They show that Qwen3.6-35B-A3B goes from 41.1 to 53.3 on Terminal-Bench 2.1 by trainin
arXiv:2610.14344v1 Announce Type: new Abstract: Scaling deep learning models efficiently requires predictable performance transfer across compute regimes. Establishing reliable hyperparameter transfer
arXiv:2610.14043v1 Announce Type: cross Abstract: Frontier language models reliably refuse to explain how to cause physical harm. We ask whether this refusal survives when the same models control a ro
arXiv:2610.17426v1 Announce Type: new Abstract: Clinical decision-making is a sequential process of information collection, clinical inference, and forecasting under partial observability, often witho
tl;dr: new qwen finetune that simultaneously pushes the frontiers of small model capability and reasoning efficiency. Hey everyone, We publish two post-trained versions of Qwen 3.8 27B called Sabi-Noe
arXiv:2610.16957v1 Announce Type: cross Abstract: Standard hate speech detection datasets often emphasize explicit harmful language, but directed harm can also be expressed indirectly, without clearly